Uh oh!
There was an error while loading. Please reload this page.
fix: fix PyTorchModel deployment crash on Windows - #1212
Merged
Conversation
Starting from version `1.2` deployment of PyTorchModel always initiates repacking when uploading the code. It is done by calling `_upload_code` with `repack=True` parameter. On Windows, this will lead to a crash, because `os.path.join` method used to generate `repacked_model_uri` will use backslashes instead of slashes. So `utils.repack_model` will fail to parse it.
sagemaker-bot
commented
Jan 7, 2020
Collaborator
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
sagemaker-bot
commented
Jan 7, 2020
Collaborator
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
sagemaker-bot
commented
Jan 7, 2020
Collaborator
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
sagemaker-bot
commented
Jan 7, 2020
Collaborator
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
laurenyu
approved these changes
Jan 7, 2020
laurenyu
left a comment
Contributor
There was a problem hiding this comment.
thanks for the contribution!
knikure added a commit
to knikure/sagemaker-python-sdk
that referenced
this pull request
Oct 26, 2023
…- Galactus (aws#1212) Co-authored-by: Gary Wang <garywan@amazon.com> Co-authored-by: Gary Wang <38331932+gwang111@users.noreply.github.com> Co-authored-by: Raymond Liu <tzujui@amazon.com> Co-authored-by: John Barboza <johnbarb@amazon.com> Co-authored-by: Malav Shastri <malavhs@amazon.com> Co-authored-by: Mufaddal Rohawala <89424143+mufaddal-rohawala@users.noreply.github.com> Co-authored-by: Mike Schneider <104035434+xncqr@users.noreply.github.com> Co-authored-by: Bhupendra Singh <bsinghtt@amazon.com> Co-authored-by: ci <ci> Co-authored-by: Malav Shastri <57682969+malav-shastri@users.noreply.github.com> Co-authored-by: Keshav Chandak <chakesh@amazon.com> Co-authored-by: Zuoyuan Huang <david.hzy97@gmail.com> Co-authored-by: evakravi <69981223+evakravi@users.noreply.github.com> Co-authored-by: Keshav Chandak <keshav.chandak1995@gmail.com> Co-authored-by: Alexander Pivovarov <apivovarov@gmail.com> Co-authored-by: SSRraymond <105880250+SSRraymond@users.noreply.github.com> Co-authored-by: Ruilian Gao <ruiliann@amazon.com> Co-authored-by: Ao Guo <aoguo@amazon.com> Co-authored-by: qidewenwhen <32910701+qidewenwhen@users.noreply.github.com> Co-authored-by: mariumof <99500633+mariumof@users.noreply.github.com> Co-authored-by: matherit <97054789+matherit@users.noreply.github.com> Co-authored-by: amzn-choeric <105388439+amzn-choeric@users.noreply.github.com> Co-authored-by: Ao Guo <72373287+aoguo64@users.noreply.github.com> Co-authored-by: Sally Seok <sallyseo@amazon.com> Co-authored-by: Erick Benitez-Ramos <141277478+benieric@users.noreply.github.com> Co-authored-by: Qingzi-Lan <83724147+Qingzi-Lan@users.noreply.github.com> Co-authored-by: Sally Seok <49303563+sallyseok@users.noreply.github.com> Co-authored-by: Manu Seth <22492939+mseth10@users.noreply.github.com> Co-authored-by: Miyoung <myoung8739@gmail.com> Co-authored-by: Sarah Castillo <67715639+secastillo@users.noreply.github.com> Co-authored-by: EC2 Default User <ec2-user@ip-172-16-36-195.us-west-2.compute.internal> Co-authored-by: EC2 Default User <ec2-user@ip-172-16-59-135.us-west-2.compute.internal> Co-authored-by: EC2 Default User <ec2-user@ip-172-16-54-104.us-west-2.compute.internal> Co-authored-by: Xin Wang <xwnamz@amazon.com> Co-authored-by: stacicho <stacicho@amazon.com> Co-authored-by: martinRenou <martin.renou@gmail.com> Co-authored-by: jiapinw <jiapinw@amazon.com> Co-authored-by: Akash Goel <goelakash93@gmail.com> Co-authored-by: Joseph Zhang <cjz@amazon.com> Co-authored-by: Harsha Reddy <116772862+sthodimi@users.noreply.github.com> Co-authored-by: Haixin Wang <98612668+haixiw@users.noreply.github.com> Co-authored-by: Kalyani Nikure <110067132+knikure@users.noreply.github.com> Co-authored-by: Xin Wang <mr.xin.wang@gmail.com> Co-authored-by: Gili Nachum <gilinachum@gmail.com> Co-authored-by: Jose Pena <josejuan-98@hotmail.com> Co-authored-by: cansun <80425164+can-sun@users.noreply.github.com> Co-authored-by: AWS-pratab <110930005+AWS-pratab@users.noreply.github.com> Co-authored-by: shenlongtang <127143389+shenlongtang@users.noreply.github.com> Co-authored-by: Zach Kimberg <zachary@kimberg.com> Co-authored-by: chrivtho-github <106935070+chrivtho-github@users.noreply.github.com> Co-authored-by: Justin <justinm088@hotmail.com> Co-authored-by: Duc Trung Le <leductrungxf@gmail.com> Co-authored-by: HappyAmazonian <91216626+HappyAmazonian@users.noreply.github.com> Co-authored-by: cj-zhang <32367995+cj-zhang@users.noreply.github.com> Co-authored-by: Matthew <117945769+mwfongAWS@users.noreply.github.com> Co-authored-by: Zach Kimberg <kimbergz@amazon.com> Co-authored-by: Rohith Nadimpally <rnadimp@amazon.com> Co-authored-by: rohithn1 <35045363+rohithn1@users.noreply.github.com> Co-authored-by: Victor Zhu <victorvctr.zh@gmail.com> Co-authored-by: jbarz1 <64875874+jbarz1@users.noreply.github.com> Co-authored-by: Mohan Gandhi <mohgan@amazon.com> Co-authored-by: Mohan Gandhi <mohan.sudhan@gmail.com> Co-authored-by: Barboza <johnbarb@88665a270f35.ant.amazon.com> Co-authored-by: ruiliann666 <141953824+ruiliann666@users.noreply.github.com> fixes (aws#963) fix: skip tensorflow local mode notebook test (aws#4060) Fix TorchTensorSer/Deser (aws#969) fix (aws#971) fix local container mode (aws#972) Fix auto detect (aws#979) Fix routing fn (aws#981) fix: tags for jumpstart model package models (aws#4061) fix: pipeline variable kms key (aws#4065) fix: jumpstart cache using sagemaker session s3 client (aws#4051) fix: gated models unsupported region (aws#4069) fix local container serialization (aws#989) fix custom serialiazation with local container. Also remove a lot of unused code (aws#994) Fix custom serialization for local container mode (aws#1000) fix pytorch version (aws#1001) Fix unit test (aws#990) Fix unit tests (aws#1018) Fix happy hf test (aws#1026) fix logic setup (aws#1034) fixes (aws#1045) Fix flake error in init (aws#1050) fix (aws#1053) fix: pipeline upsert failed to pass parallelism_config to update (aws#4066) fix: temporarily skip kmeans notebook (aws#4092) fixes (aws#1051) Fix missing absolute import error (aws#1057) Fix flake8 error in unit test (aws#1058) fixes (aws#1056) Fix flake8 error in integ test (aws#1060) Fix black format error in test_pickle_dependencies (aws#1062) Fix docstyle error under serve (aws#1065) Fix docstyle error in builder failure (aws#1066) fix black and flake8 formatting (aws#1069) Fix format error (aws#1070) Fix integ test (aws#1074) fix: HuggingFaceProcessor parameterized instance_type when image_uri is absent (aws#4072) fix: log message when sdk defaults not applied (aws#4104) fix: handle bad jumpstart default session (aws#4109) Fix the version information, whl and flake8 (aws#1085) Fix JSON serializer error (aws#1088) Fix unit test (aws#1091) fix format (aws#1103) Fix local mode predictor (aws#1107) Fix DJLPredictor (aws#1108) Fix modelbuilder unit tests (aws#1118) fixes (aws#1136) fixes (aws#1165) fixes (aws#1166) fix: auto ml integ tests and add flaky test markers (aws#4136) fix model data for JumpStartModel (aws#4135) fix: transform step unit test (aws#4151) fix: Update pipeline.py and selective_execution_config.py with small fixes (aws#1099) fix: Fixed bug in _create_training_details (aws#4141) fix: use correct line endings and s3 uris on windows (aws#4118) fix: js tagging s3 prefix (aws#4167) fix: Update Ec2 instance type to g5.4xlarge in test_huggingface_torch_distributed.py (aws#4181) fix: import error in unsupported js regions (aws#4188) fix: update local mode schema (aws#4185) fix: fix flaky Inference Recommender integration tests (aws#4156) fix: clone distribution in validate_distribution (aws#4205) Fix hyperlinks in feature_processor.scheduler parameter descriptions (aws#4208) Fix master merge formatting (aws#1186) Fix master unit tests (aws#1203) Fix djl unit tests (aws#1204) Fix merge conflicts (aws#1217) fix: fix URL links (aws#4217) fix: bump urllib3 version (aws#4223) fix: relax upper bound on urllib in local mode requirements (aws#4219) fixes (aws#1224) fix formatting (aws#1233) fix byoc unit tests (aws#1235) fix byoc unit tests (aws#1236)
Evan-W-ang added a commit
to Evan-W-ang/sagemaker-python-sdk
that referenced
this pull request
Jun 8, 2026
Starting from version `1.2` deployment of PyTorchModel always initiates repacking when uploading the code. It is done by calling `_upload_code` with `repack=True` parameter. On Windows, this will lead to a crash, because `os.path.join` method used to generate `repacked_model_uri` will use backslashes instead of slashes. So `utils.repack_model` will fail to parse it.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Starting from version
1.2deployment of PyTorchModel always initiates repacking when uploading the code. It is done by calling_upload_codewithrepack=Trueparameter.On Windows, this will lead to a crash, because
os.path.joinmethod used to generaterepacked_model_uriwill use backslashes instead of slashes. Soutils.repack_modelwill fail to parse it.Description of changes:
Replaced
os.path.join(bucket, key_prefix, "model.tar.gz")with'/'.join(...)for creating repacked_model_uri.Merge Checklist
Put an
xin the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.